On the other hand, it doesn’t really matter so much anymore.
LLM is the new search. I can ask it the actual question I have and it will give me the answer. If it’s not exactly what I need I can ask it to specify further.
Contrast that with a search engine that just gives me a ton of bookmarks to sift through to see if they actually might answer my question or are just clickbait.
Of course there’s still some times when you need search, like when you need to find an actual website, or when you need a source reference. But really the need for me is greatly reduced now.
Be careful relying on LLMs for “searching”. I’m speaking from experience here - getting actually accurate results from the current generation of LLMs, even with RAG, is difficult. You might get accurate results most of the time (even 80% or more), but it can be difficult to identify the inaccurate results due to the confidence models present their output with when hallucinating.
Also, if your LLM isn’t doing retrieval-augmented generation (RAG), then it isn’t actually a search and won’t find results more recent than the data it was trained off of.
deleted by creator
On the other hand, it doesn’t really matter so much anymore.
LLM is the new search. I can ask it the actual question I have and it will give me the answer. If it’s not exactly what I need I can ask it to specify further.
Contrast that with a search engine that just gives me a ton of bookmarks to sift through to see if they actually might answer my question or are just clickbait.
Of course there’s still some times when you need search, like when you need to find an actual website, or when you need a source reference. But really the need for me is greatly reduced now.
Be careful relying on LLMs for “searching”. I’m speaking from experience here - getting actually accurate results from the current generation of LLMs, even with RAG, is difficult. You might get accurate results most of the time (even 80% or more), but it can be difficult to identify the inaccurate results due to the confidence models present their output with when hallucinating.
Also, if your LLM isn’t doing retrieval-augmented generation (RAG), then it isn’t actually a search and won’t find results more recent than the data it was trained off of.